Development of an Extended Acute Care Surgery Service in Response to the Covid-19 Global Pandemic: Assessment of Clinical Patient Outcomes and Staff Psychological Well-being

Sachin Mathur (  sachinmathur60@gmail.com ) Khoo Teck Puat Hospital https://orcid.org/0000-0003-1795-9587 Chung Fai Jeremy Ng Singapore General Hospital Fangju Koh Singapore General Hospital Mingzhe Cai Singapore General Hospital Gautham Palaniappan Singapore General Hospital Yun Le Linn Singapore General Hospital Huiling Linda Lim Singapore General Hospital Ramu Lakshman Singapore General Hospital Xiao Shuang Ling Singapore General Hospital Sock Teng Chin Singapore General Hospital Hiang Khoon Tan Singapore General Hospital


Introduction
The COVID-19 pandemic has provided humanity with enormous healthcare and economic challenges.
Since Chinese New Year (CNY), the epicentre has shifted from China to Europe and the United States with over 4 million people infected and 250,000 deaths. The Singapore government escalated its 'Disease outbreak response system condition (DORCSCON) level to Orange on February 7 instigating containment strategies of isolating infected cases, aggressive contact tracing and widespread testing [1]. Despite mitigation efforts such as border closures, travel restrictions, enforced social distancing and a virtual lockdown on April 7, the number of infected cases has risen sharply to over 20,000 (3800/million) primarily via community spread.
It was accepted that an 'all of Government, all of society' strategy was required to combat this crisis.
Acute care surgery (ACS) involves managing trauma, emergency general surgery (GS) and critically ill patients. Unlike elective surgery, ACS cases cannot be cancelled during the pandemic and without appropriate strategies, ACS patients will compete for precious hospital resources such as acute ward beds, radiological tests and personal protective equipment (PPE). Furthermore, the safety and well-being of staff already working within this stressful sub-specialty comes into sharp focus [2].
Singapore General Hospital (SGH) is the leading public healthcare provider in Singapore and our ACS service has been in existence since 2016. This study describes the changes made to extend our system (eACS) to primarily safeguard against the risk of COVID-19 transmission to staff and patients, e ciently manage hospital resources and provide round the clock care. We also assessed our team-members attitudes towards the changes as well as their psychological well-being during this timeframe.

ACS Model
The ACS 'surgeon of the week' team is staffed by dedicated ACS and sub-specialty surgeons. The team admits all ACS patients during daytime hours from Monday-Friday with minimal con ict from elective surgery, endoscopy or clinics. After-hours and weekend care is provided by a roster of department surgeons with patients handed over to the ACS team in daily morning meetings. Junior sta ng consists of associate consultants/fellows, senior/junior residents, non-training medical o cers and house o cers (interns). Typically, the team manages 20-30 admissions per day as well as 10-15 in-house referrals to GS from other medical and surgical disciplines.

eACS Model
Staff management An eACS committee was convened after the rst COVID-19 case in Singapore on Jan 23. The eACS team began admitting patients from February 3. The primary aim of the service was to reduce exposure of surgical staff and cross-contamination of patients by COVID-19 via segregation of the department into eACS and elective. Emergency Room (ER) admissions were reviewed on the ward to limit exposure to COVID-19; only trauma activations and unstable/septic patients were seen in the ER and required donning of N95 respirator. Any ACS patient triaged as high risk for COVID-19 required full PPE: N95 respirator, hair and face shield, gown and gloves. The eACS junior staff managed ER admissions after hours and the elective team managed existing non-eACS patients with minimal interaction. Staff in clinical areas were always expected to don a surgical mask. Social distancing between team members was enforced.
The eACS service consisted of 5-teams with a rolling 1:5 24hr call system (Fig. 1). Each team was led by either an ACS surgeon or sub-specialty (Colorectal, Hepato-Biliary, Upper-Gastrointestinal) colleague who contributed to the service in 1-2-week blocks. The sub-specialty surgeons were drafted to ensure the eACS team had a full roster to provide 24/7 care. Each team was staffed by residents, medical o cers and house o cers. Inter-team transfer of sub-specialty patients was allowed. Elective surgery and outpatient clinics for these teams was cancelled. Patients discharged from the eACS service were followed up by elective teams at least 2-weeks post-discharge.
Protocols for admission to wards, operating room, endoscopy and radiology Patients were admitted from the ER to an existing ACS ward to segregate them from elective patients. COVID-19 suspected cases were admitted to the Infectious disease (ID) isolation ward. Separate acute respiratory infection 'ARI' beds were made available in multiple wards to admit surgical patients with fever, respiratory symptoms or signs of pneumonia on CXR or CT scan. In the isolation/ARI wards a combination of full registration of staff visiting and leaving the ward was required as well as donning full PPE: N95 respirator, face and hair shield, gown and gloves.
A new work ow was established for patients who required surgery (Fig. 2). An e-consent process for suspect COVID-19 cases was established with the department of Anesthesia utilising a downloaded consent form on a Toughbook© placed in a Ziplock airtight bag. COVID-19 suspect patients required transportation by designated porters in separate elevators. An operating room (OR) with high e ciency particulate air lter (HEPA) was separated from the main OR block to reduce cross-transmission between staff/patients. All staff were to remain outside of OR during intubation and expected to don N95 respirator, face-shield and eye goggles during surgery. Low risk patients (no recent travel, contact with COVID-19 patient or fever/respiratory symptoms) underwent surgery in the main OR however anesthetic staff-maintained use of N95 respirators for intubation with standard PPE for surgeons. The protocol was tested with an in situ simulation tracking the progress of a COVID-19 positive bleeding ulcer patient with haemorrhagic shock and subsequent PEA arrest.
Stable patients that could undergo endoscopy in the outpatient setting were discharged. Urgent non-COVID-19 eACS cases were performed in the endoscopy centre with use of PPE including N95 respirator, face-shield, gown and gloves for both upper and lower GI endoscopy based upon recommendations from in-house Gastroenterologists. This has subsequently been endorsed by the American Gastroenterological Association to protect against aerosolization of COVID-19 [3]. Suspect COVID-19 patients or potentially unstable GI bleed patients underwent endoscopy in the established COVID-19 OR with full PPE.
Patients requiring urgent CT scans from the ER were transported regardless of COVID-19 status with appropriate precautions for all staff (N95 respirators). Beyond that patient work ow was determined by ward. Those in COVID-19 isolation/ARI beds were prioritised for daytime scans with protocols in place for transportation, nursing/medical escort, radiology staff PPE and necessary decontamination procedures post scan. Similar to suspect COVID-19 patients requiring OR, the patients' route from ward to CT scan was cleared and subsequently decontaminated by designated cleaning staff. Non-urgent CT scan requests were performed after 2 swab COVID-19 clearance as per ID protocol.

ACS team meetings
All ACS meetings including journal club, trauma audit, morbidity and mortality and medical student lectures and ward-based tutorials were cancelled as per senior management directives for meetings of 10 or more people.

Patients
The study was reviewed and approved by the local Institutional Review Board under waiver of consent as anonymised data was utilised. A retrospective study was performed comparing 2-month cohorts: Clinical and e ciency outcomes Prospective data collected for eACS included total number of patients seen, total in-house referrals to GS and ward round times. Data obtained from an historic time-in motion study of 10 consecutive ACS ward rounds in 2018 was used as comparison. Retrospective data obtained included demographics, surgical diagnosis and procedure (operative and endoscopy). Length of stay (LOS) was de ned as duration of time between day of admission and discharge. Further time divisions identi ed were time from ER arrival to admission, time from CT request to CT performed for patients undergoing surgery, time from CT scan to surgery start time and time from surgery until discharge. The number of annualised bed-days saved was calculated as ((eACS mean LOS -ACS mean LOS)*number of eACS cases in 2 months)*6. Financial savings made were calculated as: number of cases*change in LOS*average ward charge and daily treatment fee. Post-operative morbidity was classi ed as per Clavien-Dindo [4]. Mortality data was obtained. The number of suspected COVID-19 patients was assessed as well as the total number of COVID-19 positive patients treated in SGH during the eACS time.

Survey of team satisfaction with eACS and concerns regarding COVID-19
A survey (Qualtrics, Provo. UT) was sent to all eACS staff at the end of March and appropriate consent obtained from each participant. The survey used a 5-point Likert scale to assess 3 key domains of experience: impression of eACS compared to ACS, overall concerns regarding COVID-19 and psychological wellbeing/burnout for the preceding two months. It was adapted from the Maslach Burnout Inventory which has been utilised in multiple clinical settings [5,6] Statistical Analysis Statistical analysis was performed using IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp. Data was analysed utilising t-test, Mann-Whitney U and Chi-squared tests of comparison as appropriate. All tests were two-sided and P < 0.05 was considered statistically signi cant.

Results
Clinical and e ciency outcomes Overall, 1436 patients were admitted during both time periods (ACS 734, eACS 702). Table 1 shows the demographics. There were no healthcare worker or patient cross-infection of COVID-19. There was a signi cant reduction in patients over 50 admitted to the eACS service (403 vs 469, P = 0.012). Similar numbers of the common-most conditions were admitted as well as operative procedures performed. There were 293 GS and sub-specialty referrals seen during the 2 months eACS. There was a decrease in total endoscopies performed (90 vs 121 P = 0.03). Overall post-operative morbidity was similar in both groups however there was a trend towards reduced mortality in the eACS era (6 vs 12 cases). The number of suspect COVID-19 cases was 96 or 14% of the admitted population. Overall SGH had treated 800 COVID-19 positive cases at the time of writing.  Figure 3 shows the ward distribution for admitted patients. There was a 29% increase in patients admitted to the surgical wards during eACS (overall 75% vs 58%, P < 0.001). Of the eACS patients 14% were admitted to ARI/Isolation wards. Compared to the historical control of 86 minutes, eACS average ward rounds across 5-teams were 35% shorter (56 minutes).

Survey of team satisfaction with eACS and concerns regarding COVID-19
The survey was answered by 90/92 members of the eACS team (98% response rate).
When comparing eACS to ACS, the team members enjoyed working within the new system (Fig. 4a). Speci cally, the majority felt they were able to maintain rostered junior hours and focus on patient care. They also felt the service provided timely access to scans and surgery for eACS patients.
With regard to COVID-19, there was a clear pattern of concern amongst staff that they were at risk of contracting the disease and passing it onto their family members (Fig. 4b). Conversely, there was near universally acceptance that the system provides appropriate protections and allows for segregation of healthcare workers.
The speci c questions and answers for burnout/psychological wellbeing are shown in Fig. 4c. The vast majority of respondents felt appreciated, that they were positively in uencing patients' lives and they were accomplishing worthwhile things at work. Questions pertaining to frustration with work or feeling emotionally drained or fatigued were met with almost universal disagreement.

Discussion
This study describes the rapid development of an eACS service in response to the COVID-19 outbreak. There were no team-member or patient cross-infections from COVID-19 during the study period. The eACS model resulted in improved e ciency of care, clinical outcomes and reduced costs compared to the previous ACS service. Team morale and satisfaction with the service remained high despite concerns about contracting COVID-19 or passing it on to family members.

The need for an eACS model (preparedness)
After the SARS outbreak of 2003 where 5 healthcare workers lost their lives in Singapore, preparedness models were created for future outbreaks [7]. Senior management dictated a zero tolerance for healthcare worker infection. By separating eACS and elective streams within the department and adapting the service to 5 teams, we prevented its collapse secondary to individual infection or quarantine. A similar approach was utilised by Ngoi et al in managing Oncology patients by adapting their department into 2 separate teams [8]. Further limitations such as restricted access to ER and inter-team meetings/social gatherings facilitated this result. Furthermore, by reducing costs we could potentially divert vital resources towards maintaining PPE/ventilators for those on the front-line.
Early co-ordination with the ID, ER, OR, porters and cleaning departments were able to protocolise the movement of suspect COVID-19 patients to limit staff exposure. This has been reinforced in recent global surgical guidelines for COVID-19 [9]. Protocols for PPE utilisation were developed to minimise wastage of resources. The use of in situ simulation enabled aspects of team dynamics and protocols to be tested in a safe environment and has been used extensively in surgical settings [10,11]. We were able to modify work ows surrounding blood transfusion requests, patient transportation and time required to setup COVID-19 OR. Furthermore the establishment of a designated leader (OR Anesthetist) to streamline communication between staff members and co-ordinate the logistics of transfer was encouraged with the aim of reducing the time from OR activation until patient arrival to 20 minutes [12]. Ross et al in their multi-tiered response to COVID-19 placed the ACS team front and centre with ACS staff transitioned to ICU whereas emergency GS and trauma cases were handled by suitably trained faculty [13]. In times of overwhelming COVID-19 admissions, it would be an appropriate consideration in Singapore.
eACS clinical and e ciency outcomes ACS models of care have consistently shown improved outcomes including LOS, costs, reduced complications and mortality [14][15][16][17][18]. Our eACS model reduced LOS further with signi cant cost savings to the institution in just 2-months. Furthermore, the bed-savings made would assist with preparing wards for the expected COVID-19 surge. The decreased LOS was achieved through multiple pathways: 1.
Reduced admission times for surgical patients in the ER will allow for more beds for COVID-19 suspect patients.

2.
The dedicated ACS ward (not present in 2019) provided a natural home for admitted patients, with nurses and ancillary medical staff attuned to their needs. The reduced ward round times highlighted the e ciencies that can be achieved.

3.
Sequestering suspect COVID-19 patients to ARI/isolation wards expedited care for these patients who were prioritised for swab clearance, CT scans and OR if required. 4.
Rapidly established work ow with radiology and OR/Anesthesia was achieved highlighted by the reductions in time from CT scan to surgery and from surgery to discharge for eACS patients.

5.
Our laboratory partners developed a novel COVID-19 test which enabled accelerated clearance of suspect patients within hours. It is a reverse transcription real time polymerase chain reaction (RT-PCR) targeting the E-gene of COVID-19. The assay was developed based upon published protocols [19].
The overall workload between both time-periods was similar but also consistent for the downward uctuation that occurs in February because of CNY. Fewer older admissions were noted and likely related to government directives to stay at home. Fewer endoscopy cases related to stable patients being offered outpatient evaluation. Reduction in trauma cases were noted which may be related to a combination of government DORSCON orange status, stay at home mandates and lockdown. One aspect of work ow which did increase was inter-team transfers and reviews by the sub-specialist surgeons. This was expected since these reviews were not seen by elective teams anymore. The junior manpower of the eACS service was reduced from February to March to re ect the reduction in admissions noted.

Staff satisfaction and morale
The eACS service was well received by the vast majority of the team. The structure of the service leads to 'off-days' during which personal time to de-stress appears to be helping. Whilst not a complete picture of the mental state of the team, the results are encouraging given that up to 61% of US surgical residents exhibited burnout on at least one of three sub-scales: emotional exhaustion, depersonalisation and personal accomplishment [20].
It is clear that concerns for personal safety and the health of loved ones is prevalent amongst the teammembers. However, the survey also shows that they felt the system prioritised their safety, which was one of our key goals. This has likely contributed to a positive psychological well-being as assessed in the survey. An appropriate level of PPE is vital and out of the control of most clinicians hence reliance upon the institution and government will always be there.
Support systems are imperative for healthcare workers both at home and at the institution level. SGH has multiple links for staff to access psychological support mechanisms in-house. The World Health Organisation has advocated for de-stigmatising healthcare workers from abuse in the community, encouraging workers to avoid unhelpful coping strategies and remembering that it's a marathon not a sprint [21].
Limitations A limitation of this study is the generalisability of the eACS model. When consistent attempts to ' atten the curve' are made, this eACS strategy may assist to limit staff and patient exposure to COVID-19. However, for our heroic colleagues working in healthcare systems decimated by COVID-19, they will have their own unique strategies to deal with ACS cases and we look forward to learning from them. This pandemic is far from over and adaptability of systems will be the key to success. Our strategy will need to be re-examined in future studies beyond COVID-19.

Conclusions
This is the rst study to report results after the implementation of an extended ACS model of care in response to the COVID-19 global pandemic. We have shown improved e ciency of care and clinical outcomes for these patients. Furthermore this study uniquely addresses the psychological well-being of our team-members in response to the pandemic and highlights the need for robust systems to protect workers/patients from cross-infection.  Figure 1 Enhanced ACS service from February 3, 2020 a Survey responses to questions differentiating ACS and eACS systems b Survey responses to questions regarding COVID-19 concerns and safety at work c Survey responses to questions speci cally related to burnout and psychological well-being